Re: [SeaBIOS] [coreboot] Seabios problems

Do you have an idea to bypass the dma problem you tell me ?
> Message du 17/01/10 à 18h43
> De : "Kevin O'Connor" 
> A : [e-mail]
> Copie à : [e-mail], "coreboot" 
> Objet : Re: [coreboot] Seabios problems
>
> On Sun, Jan 17, 2010 at 09:29:08AM +0100, [e-mail] wrote:
> > I am back.
> > I give you those logs:
> > I still have keyboard troubles in debuglevel=4
> > Vista doesn't boot with your patch.
> >
> > Thanks for your help.
>
> Thanks.  Can you try the following for the keyboard problem?
>
> --- a/src/ps2port.c
> +++ b/src/ps2port.c
> @@ -306,6 +306,7 @@ aux_command(int command, u8 *param)
>  static void
>  process_ps2irq(void)
>  {
> return;
> u8 status = inb(PORT_PS2_STATUS);
> if (!(status & I8042_STR_OBF)) {
>     dprintf(1, "ps2 irq but no data.\n");
>
> Your keyboard definitely wont work after this, but I'd like to see if
> the init succeeds with it.  (I think there may be a case where
> interrupts are messing with the initialization.)
>
> -Kevin

____________________________________________________

Je m’évite la cohue dans les magasins et je profite des petits prix sur http://shopping.voila.fr

[coreboot] Seabios problems

[Kevin O'Connor]

On Mon, Jan 18, 2010 at 06:14:51PM +0100, [e-mail] wrote:
> Do you have an idea to bypass the dma problem you tell me ?

The latest SeaBIOS git has ATA DMA disabled by default now.

-Kevin

[coreboot] Seabios problems

[Kevin O'Connor]

On Mon, Jan 18, 2010 at 06:14:51PM +0100, [e-mail] wrote:
> > Message du 17/01/10 à 18h43
> > De : "Kevin O'Connor" 
> > --- a/src/ps2port.c
> > +++ b/src/ps2port.c
> > @@ -306,6 +306,7 @@ aux_command(int command, u8 *param)
> >  static void
> >  process_ps2irq(void)
> >  {
> > return;
> > u8 status = inb(PORT_PS2_STATUS);
> > if (!(status & I8042_STR_OBF)) {
> >     dprintf(1, "ps2 irq but no data.\n");
> >
> > Your keyboard definitely wont work after this, but I'd like to see if
> > the init succeeds with it.  (I think there may be a case where
> > interrupts are messing with the initialization.)

init ps2port
/37fce000\ Start thread
|37fce000| i8042 flushed aa (status=11)
|37fce000| i8042 flushed 7e (status=11)
|37fce000| i8042 flushed f0 (status=11)
|37fce000| i8042 flushed 7e (status=11)
|37fce000| i8042 flushed 7e (status=11)
|37fce000| i8042 flushed f0 (status=11)
|37fce000| i8042 flushed 7e (status=11)
|37fce000| ps2_recvbyte timeout
|37fce000| keyboard initialized
\37fce000/ End thread

So, it successfully initialized the keyboard this time.  Can you
confirm that it reliably reports "keyboard initialized" now instead of
"keyboard self test failed"?

I'll put together a patch to make sure interrupts don't mess with the
init process.

Thanks,
-Kevin